RESEARCH ARTICLE Century-Old DNA Barcodes Reveal Phylogenetic Placement of the Extinct Jamaican Sunset , sloanus Cramer (: )

Vazrick Nazari1*, B. Christian Schmidt1, Sean Prosser2, Paul D. N. Hebert2

1 Canadian National Collection of , Arachnids and Nematodes, Ottawa Research and Development a11111 Centre, Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada, 2 Centre for Biodiversity Genomics, Biodiversity Institute of Ontario, University of Guelph, Guelph, Ontario, Canada

* [email protected]

Abstract OPEN ACCESS Analysis of the DNA barcode region of the cytochrome c oxidase 1 gene from a specimen Citation: Nazari V, Schmidt BC, Prosser S, Hebert of the extinct Jamaican sunset moth, Urania sloanus, places this species as a sister to the PDN (2016) Century-Old DNA Barcodes Reveal Central American U. fulgens. We found that all Urania F. species were closely related Phylogenetic Placement of the Extinct Jamaican Sunset Moth, Urania sloanus Cramer (Lepidoptera: (<2.8% maximum divergence at COI), with the Cuban endemic U. boisduvalii appearing as Uraniidae). PLoS ONE 11(10): e0164405. sister to the rest. The low divergence in DNA barcodes and genitalic structures indicate that doi:10.1371/journal.pone.0164405 the Cuban U. poeyi and eastern Brazilian U. brasiliensis are geographic segregates of U. Editor: Dorothee Huchon, Tel Aviv University, fulgens and U. leilus respectively, so the former two taxa are accordingly recognized as ISRAEL subspecies. Received: June 7, 2016

Accepted: September 23, 2016

Published: October 20, 2016

Copyright: © 2016 Nazari et al. This is an open access article distributed under the terms of the Introduction Creative Commons Attribution License, which permits unrestricted use, distribution, and The Jamaican Sunset moth, Urania sloanus (Cramer, 1776), often regarded as one of the most reproduction in any medium, provided the original beautiful of all , became extinct just 116 years after its description [1,2]. Endemic to author and source are credited. Jamaica, it inhabited low-elevation tropical rainforest containing its host plant L. Data Availability Statement: All new sequences (), especially O. triandra L. (Jamaican Cobnut) and O. diandra L. [3–5]. Urania are now deposited on GenBank (accession sloanus was named by Dutch entomologist Pieter Cramer [6] in honor of Sir Hans Sloane numbers KX781954-KX781991). Additional (1660–1753), an English naturalist and collector who explored Jamaica from 1687 to 1689, sequences used in Supporting Info S1. DOI for the recording and illustrating the species for the first time in his travel diaries published in 1725 dataset: dx.doi.org/10.5883/DS-URANIA. [7]. The life history and behaviour of U. sloanus was subsequently described in detail by Philip Funding: This study was supported by grants to Gosse [3,8–9], who reported that adults were common from mid-March to late July but usually Paul D.N. Hebert (PDNH) from Natural Sciences rare in the winter. Periodic episodes of abundance, when swarms of these moths occurred at and Engineering Research Council of Canada flowering trees in the Blue Mountains of Portland, were intervened by years of great scarcity (NSERC) and from the Government of Canada through Genome Canada and the Ontario [1,8,10–11]. The moth became increasingly rare in the early 1890s; the last sighting occurred in Genomics Institute in support of the International 1894–1895 [1], although the species may have survived until 1908 [12]. Lees and Smith [11] Barcode of Life Project. postulated that populations of U. sloanus declined below a sustainable level due to the loss of

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 1 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

Competing Interests: The authors have declared its main larval host through habitat destruction linked to agriculture or hurricanes. Vinciguerra that no competing interests exist. [12] suggested the species may still persist in remote areas with appropriate habitat and host plants. In addition to U. sloanus, five other species of Urania Fabricius are generally recognized: U. leilus (Linnaeus, 1758), widespread in tropical South America east of the ; U. fulgens Walker, 1854, a migratory species distributed from Mexico through Central America to north- western South America (occasionally straying to Florida and Texas [13] as well as Jamaica [Lees, pers. comm.]), and U. boisduvalii Guérin-Meneville, 1829, endemic to Cuba. The status of two other species, namely U. brasiliensis (Swainson, 1833) (Atlantic coast of ) and U. poeyi (Herrich-Schäffer, 1866) (Cuba) is controversial as they have been treated either as sepa- rate species [11,14–17], or subspecies or synonyms of U. fulgens and U. leilus, respectively [18]. Urania fulgens can exhibit forms that are externally very similar to U. leilus, and these two taxa have also been suggested to be conspecific [19]. By contrast, only U. sloanus shows pink and red coloration on the dorsal hindwing. The diagnostic value of genitalic structures has not been evaluated in Urania. In addition to the Neotropical Urania, the pantropical subfamily contains six other genera ( Hübner, Hübner, Cyphura Warren, Hübner, Urap- teritra Viette and Urapteroides Moore) with many large, tailed diurnal species with a spectacu- lar phenotype. The family Uraniidae apparently originated in the late Cretaceous, ~75 Mya [20] from a common ancestor with the Sematuridae [21–22] although no comprehensive phy- logeny is available for it. Lees and Smith [11] provided a preliminary phylogeny for the Uranii- nae based on seven larval and adult tympanic characters, while molecular studies [20–22] have incorporated a few representative uraniid genera within broader lepidopteran phylogenies. Urania and Chrysiridia were resolved as sister genera by Lees and Smith [11] based on shared larval and tympanic morphology and further supported by the diurnality of their adults, wing scale iridescence, and their shared use of Omphalea as larval hosts. A combined molecular and morphological analysis [22], however, found weak support for Alcides as the sister group to Urania, with both sisters to Chrysiridia. Urania moths are unique among Lepidoptera in having stridulatory organs on their male prothoracic legs [23–25]. They are also among the very few moths that form migrant popula- tions, being chemically defended against predators by polyhydroxy alkaloids sequestered from their host plants, Omphalea [26–27]. Species in the same genus are also hosts for Afrotropical Chrysiridia, as well as the Indo-Australian Alcides and Lyssa [11,28]. In fact, the disjunct distri- butions of these moths almost exactly correspond to the global range of Omphalea, suggesting a deep historical relationship, perhaps even predating the final break-up of Gondwanaland in the Cretaceous, ca. 100–65 Mya [11]. All Urania species are allo- or parapatric with respect to each other and show island ende- mism (U. boisduvalii, U. poeyi, U. sloanus), but species relationships have never been studied. Here we investigate the pattern of sequence divergence at COI among species of Urania, including the extinct U. sloanus, in the large context of family Uraniidae.

Methods and Materials Samples 40 specimens from all six recognized species of Urania, including one specimen of U. sloanus, were sequenced for COI. In addition, to aid phylogenetic inferences we included 14 representa- tives from the other six genera of Uraniinae (7), subfamilies Auzeinae (2), (2) and Microniinae (2), as well as family Sematuridae (1), recognized as sister to Uraniidae [21]. Sequences were downloaded from GenBank or BOLD with permission (S1 Table). Genitalia

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 2 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

were prepared following the methods of Lafontaine [29]. Male genitalia of all Urania taxa except U. sloanus were examined (n = 5); females were not available for all species. Cleaned, stained genitalia were stored and examined in 30% ethanol, and slide-mounted in Euparal before being photographed using a Leica DFC450 digital camera and the Leica Application Suite 4.7.1 software package.

Amplification and sequence analysis A single leg was removed from each specimen and sent to Biodiversity Institute of Ontario in Guelph, Canada for DNA extraction, amplification, and sequencing. Non-destructive DNA extraction was used for the leg from U. sloanus. All work was performed in a dedicated clean lab using UV sterilized equipment to minimize the risk of contamination. DNA extraction and purification was performed following the methods of Prosser et al [30] and the 658 bp COI bar- code region was amplified for both Sanger-based sequencing and next-generation sequencing (NGS). For Sanger sequencing, non-tailed primers (S2 Table) were used to amplify six short, overlapping fragments using a modified version of the method described in Lees et al [31]. Each of the six fragments from each sample were bi-directionally sequenced on an ABI 3730XL capillary sequencer (Applied Biosystems) and the resulting traces were edited and assembled into full contigs using CodonCodev. 3.0.1 (CodonCodeCorporation). Because the failure of any of the six fragments to amplify or sequence resulted in missing barcode data (S3 Table), we also sequenced each sample via NGS using the method described in Prosser et al [30]. Briefly, instead of amplifying six overlapping amplicons in separate reac- tions, the same primers were multiplexed in two PCR reactions per sample, and the resulting amplicons from each sample were pooled and sequenced on an Ion Torrent PGM (Thermo Fisher) using a 318 v2 chip and OT2 400bp chemistry (Thermo Fisher). Approximately 4.2M sequence reads were generated, which were demultiplexed into an average of 353K (±143K) reads per specimen. For each specimen, reads were processed to remove primer sequences (Cutadapt v1.8.1) and then filtered based on a minimum quality score of QV20 and a mini- mum length of 100bp (Sickle v1.33). The reads were then dereplicated (Fastx Toolkit v0.0.14) and assembled into a full-length barcode contig by aligning to a reference barcode sequence [32]. The resulting NGS sequences were merged with their Sanger-based counterparts to fill in the missing data (S4 Table). Final sequences were deposited in GenBank (accessions XXX-XXX)and all NGS reads were deposited in the Sequence Read Archive (accessions XXX-XXX).All of the barcode records are also publicly available in the BOLD dataset “DS-UR- ANIA”, accessed at dx.doi.org/XX.XXXX/DS-URANIA.

Phylogenetic analysis, statistical parsimony network, and ancestral area reconstruction Uncorrected Kimura 2 Parameter (K2P) distances for COI barcodes were calculated using modules implemented in BOLD, and subsequently summarized in Excel (Table 1). A Neigh- bour-Joining tree was also constructed in BOLD under a K2P distance model. Bootstrap values for 1000 replicates were calculated in MEGA 6.06 [33]. To infer a phylogenetic hypothesis for Urania, a Likelihood analysis was carried out with PhyML [34] using AIC criterion with boot- strap of 100 replicates and under GTR + Γ + I + F model as selected by the automatic model- selection module implemented in PhyML. We also conducted a Bayesian analysis in MrBayes v.3.2. [35] under the GTR + Γ + I model with two simulations, independent Markov chain Montecarlo (MCMC) runs starting from different random trees, each with 3 heated chains and 1 cold chain for 50 million generations.

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 3 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

Table 1. Uncorrected p-distances of the mtDNA barcode sequences between Urania species. U. boisduvalii U. fulgens U. f. poeyi U. leilus U. l. brasiliensis U. sloanus Urania boisduvalii (n = 3) 0.10 ± 0.1 Urania fulgens (n = 15) 2.78 ± 0.1 0.1 ± 0.1 Urania fulgens ssp.poeyi (n = 2) 2.53 ± 0.5 0.67 ± 0.4 0.65 (n = 8) 2.54 ± 0.1 2.00 ± 0.1 2.69 ± 0.5 0.04 ± 0.1 Urania leilus ssp. brasiliensis (n = 1) 2.55 ± 0.1 1.94 ± 0.1 2.57 ± 0.6 0.02 ± 0.1 - Urania sloanus (n = 1) 2.39 ± 0.1 1.15 ± 0.1 1.92 ± 0.6 1.44 ± 0.1 1.39 - doi:10.1371/journal.pone.0164405.t001

Statistical Parsimony Network was constructed in TCS 1.21 with a 90% confidence limit for parsimony [36]. Only full-length (658 bp) COI barcodes were included in the analysis, and with the exception of two U. poeyi sequences, all those with ambiguous bases were excluded. Dispersal-vicariance optimization of ancestral areas implemented in DIVA 1.1 [37] was used to infer the biogeographic history of the group using five biogeographic units, as follows: SA: South America, CA: Central America, C: Cuba, J: Jamaica, and M) Madagascar. The least num- ber of dispersals (11) was achieved when maxareas was set to 1.

Results Male genitalic structure is relatively conserved among Urania with the most pronounced dif- ferences occurring in the shape and armature of the vesica. We could find no consistent differ- ences in structure between members of two species pairs (U. fulgens vs. U. poeyi, U. leilus vs. U. brasiliensis), but U. fulgens and U. leilus differ in the size and position of diverticula in the vesica, and notably in the absence of the left lateral spine field in U. fulgens. There are no obvi- ous differences in valve structure between the latter two species, although overall shape may differ subtly. In contrast, U. boisduvalii differed in vesica structure and possessed a more rounded, elongate valve shape than U. fulgens and U. leilus. Our ancient DNA technique [30] produced a full-length DNA barcode (658bp) for Urania sloanus and several other species, and fragments of various lengths for other old specimens. Throughout Bayesian, ML (not shown) and NJ analyses, Urania formed a highly supported monophyletic group with the same topology, with U. boisduvalii, U. leilus+ U. brasiliensis, U. sloanus, and U. fulgens + U. poeyi branching off successively. As expected with mitochondrial data, higher relationships within the subfamily (not of concern in this study) were not sup- ported and the positions of the non-Urania taxa varied across our analyses. With Coronidia (Sematuridae) used to root the trees, the two representatives of Epipleminae and Microniinae formed monophyletic clades, while Auzeiinae and Uraniinae did not. Within Uraniinae, NJ analysis placed larger species in Lyssa, Chrysiridia, Alcides and Urania closer together while the partly nocturnal, possibly non-migratory neighbouring taxa Urapteritra, Urapteroides and Cyphura stayed well outside the remaining genera. These three genera have several other differ- ences from the remaining Uraniinae (smaller size, wing shape, patterning, hind wing venation etc. [11]). Barcode data also could not resolve the sister-group relationship to Urania, however Alcides appeared to be most similar to Urania with no support, while Chrysiridia stayed further off in our Bayesian (Fig 1) and ML (not shown) analyses. Within the monophyletic genus Urania, U. sloanus appeared as sister to but distinct from the Central American species U. fulgens with moderate support. The three specimens of brasi- liensis included in our study shared identical haplotypes with U. leilus, suggesting this taxon is not distinct. Similarly, the two specimens of U. poeyi were very similar to the Central American U. fulgens, although they were different at two positions from one another.

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 4 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

Fig 1. a) Bayesian inference with posterior probabilities, and b) Neighbor-Joining tree with bootstrap support values for Uraniidae COI-5P barcode data. doi:10.1371/journal.pone.0164405.g001

Discussion The subfamily Uraniinae contains some of the most spectacular Lepidoptera with large and diurnal species whose hindwing tails and contrasting black and iridescent colours often cause them to be mistaken for . With representatives in the Neotropics, Afrotropics, Orien- tal and Australian regions, the subfamily shows a typical Gondwanan pan-tropical distribution, corresponding to that of its host plant Omphalea [11]. Gondwana began to fragment in the early Jurassic (~180 Mya), and by 110–100 Mya the continents were completely separated [38]. Presuming a vicariance scenario, the split in range of the last common ancestor of Urania must have occurred at least 100 Mya in mid- to late Cretaceous, during the final separation of South America and Africa [11]. However, these dates are far before the estimated divergence time for Chrysirida and Urania (14.1–35.8 Mya) or even the origin of the Uraniidae at 57.3–81.9 Mya [20]. The very low mtDNA differentiation (0.8%) between the two species of Chrysiridia in Africa and Madagascar, which were last connected around 165 Mya, also strongly supports one or more recent dispersal events across the Mozambique Channel. The well-documented migra- tory behaviour and long-distance mass migrations of many species of Uraniinae, including Urania [23,39–43] and Chrysiridia [44,45] provide an explanation of the more recent dispersal events in these moths.

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 5 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

Fig 2. Distribution of examined material and phylogeny of genus Urania. Values below nodes are posterior probabilities calculated in BEAST, above nodes are ancestral distributions as predicted by DIVA (SA: South America, CA: Central America, C: Cuba, J: Jamaica, and M) Madagascar). doi:10.1371/journal.pone.0164405.g002

Dispersal events must also be considered as the likely origin of the Neotropical precursor of Urania. The last common ancestor of all Urania moths was likely isolated in Cuba and South America in the Early Miocene, where it gave rise to U. boisduvalii in Cuba (Fig 2). Numerous larval and behavioural autapomorphies have been previously noted for U. boisduvalii

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 6 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

compared to both U. fulgens and U. leilus [19]. Identical genitalia and mtDNA sequences sup- port the presence of only one (U. leilus) rather than two species in South America, however considering the current geographic isolation of these populations, we maintain brasiliensis as a subspecies of U. leilus. The South American lineage dispersed to Central America likely after the formation of the Isthmus of Panama during the Middle Miocene, around 15–11 Mya [46]. The Andes began a rapid uplift in what is now and about 5 Mya, reaching today’s elevations at 2.7 Mya [47]. Genesis of the Andes likely created a natural barrier within the range of ancestral Urania, and may have separated ancestral populations of U. leilus to the southeast of the Andes and U. fulgens to the west. The taxon poeyi seems to be a recent addition to the Cuban fauna as a result of dispersal of ancestral U. fulgens from Central America, but its geographical isolation from fulgens warrants its recognition as a subspecies. The ancestral U. sloanus also seems to have been isolated in Jamaica following an earlier dispersal of an ancestral lineage of U. fulgens from Central America. These dispersal events are unlikely to be linked to the geological history of the Caribbean since all Urania seem to have diversified relatively recently as evidenced by their low mtDNA divergences (maximum 2.78% between U. bosiduva- lii & U. fulgens; Table 1), and may be better explained as the result of migratory events aided by the prevailing easterly trade winds [48]. Our molecular analyses do not support the position of the three reportedly nocturnal, non- migratory genera (Urapteritra, Urapteroides and Cyphura) within the Uraniinae, and contrary to Viette’s (1972) interpretations based on wing venation, suggests that Urapteritra and Cyphura might be sister taxa [49]. Although there is no phylogenetic support from COI-5P for any topology among these three genera, our results seem to hint at a closer affinity between Urania and Alcides rather than Chrysiridia as suggested previously [11,44] and consistent with Heikkilä et al’s [22] results based on combined morphological and molecular data. Urania, Alcides and Chrysiridia also share synapomorphies with respect to their tympanic organ [11]. Further evidence is required to clarify a phylogenetic hypothesis between these three genera.

Taxonomic considerations The validity of U. sloanus as a distinct species has never been questioned, and its 1.15% barcode divergence relative to U. fulgens support its status as a young though valid species. Urania ful- gens and U. leilus have previously been treated as being closely related [18] or even conspecific [19]. However, our molecular and morphological studies indicate that they are not sister spe- cies, since U. fulgens is more closely related to U. sloanus than to U. leilus. The Andes appear to have been an important dispersal barrier with U. fulgens occurring on the western slope and north through Central America, while U. leilus occurs from the east slope and throughout the Amazon Basin, and possibly disjunct on the Atlantic coast of Brazil. Urania poeyi has been treated as a species endemic to Cuba that is questionably distinct from U. fulgens since the two taxa are very similar in phenotype and larval appearance [11] and share the familial traits of strong sexual dimorphism and migratory behaviour. Their identical male genitalia and lack of mtDNA differentiation further supports their conspecificity. The range of U. fulgens therefore includes Cuba, either through recent or ongoing dispersal involv- ing populations from Central America, a conclusion supported by the occasional intrusion of this species into Florida and Jamaica [50]. Similarly, both barcode haplotypes detected in U. brasiliensis were also found in U. leilus (Fig 3), and there were no discernible differences in genitalic structures between these taxa (Fig 4). The high degree of similarity in morphology, genitalia and mtDNA between U. brasiliensis specimens from easternmost Brazil and the Amazonian U. leilus indicates ongoing or very recent genetic interchange, so we conclude that they are conspecific as suggested by Smith [19].

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 7 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

Fig 3. Statistical Parsimony Network of full-length Urania sequences examined in this study. Colors correspond to species in Fig 2. doi:10.1371/journal.pone.0164405.g003

Fig 4. Male genitalia of Urania species: A) Urania boisduvalii CNCGEO10060, B) Urania leilus CNCGEO10062, C) Urania fulgens ssp. poeyi CNCGEO10061. Genitalia of Urania leilus ssp. brasiliensis (CNCGEO10059) and U. fulgens (CNCGEO10058) were identical to B and C respectively and thus not shown. doi:10.1371/journal.pone.0164405.g004

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 8 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

Conclusion Despite the complex and much debated geological history of the Caribbean [51], the evolution of Urania in the region seems relatively recent and mostly based on dispersal and subsequent isolation. Island biogeographic theory predicts that most island species are likely to be recent derivatives from mainland populations, and with some exceptions, islands (particularly small islands, here not the case for Cuba) tend to be home not to ancient endemics, but rather to recent offshoots from mainland progenitors. This has important implications for conservation biology, considering that many island populations are threatened. While the loss of island line- ages is regrettable, it is important to recognize that closely allied populations often occur on neighbouring mainlands. Understanding the evolutionary relationships between island species and their mainland counterparts is thus a key consideration in island conservation efforts. The two base pairs by which U. fulgens ssp. poeyi is distinct may also flag its conservation signifi- cance in Cuba. A similar argument can be made for the anthropogenic islands of the Atlantic forest of Brazil: Only one minute Atlantic forest remnant protects both U. leilus ssp. brasiliensis and its hostplant, Omphalea brasiliensis (D. Lees, pers. comm.). Future studies on Urania should focus on obtaining full mitochondrial or nuclear genomes for all species, including U. sloanus but also other Uraniinae, to enable a more comprehensive evolutionary and divergence time analyses.

Revised classification of Urania Urania Fabricius, 1807 Urania Fabricius, 1807; Magazin f. Insektenk. (Illiger) 6: 279. Type-species: Papilio leilus Linnaeus, 1758. = Urania [Illiger], 1807; Allgem. Lit. Ztg. Halle [Jena] 1807 (No. 2): 1181 (suppr.). Type-spe- cies: Papilio leilaria Hübner, [1807]. = Lars Hübner, [1807]; Samml. exot. Schmett. 1: pl. [200]. Type-species: Papilio leilaria Hübner, [1807]. = Cydrus Billberg, 1820; Enum. Ins. Mus. Billb.: 81. Type-species: Papilio leilus Linnaeus, 1758. = Cydimon Dalman, 1825; K. VetenskAcad. Handl. 1824 (2): 407. Unnecessary replacement name for Urania Fabricius, 1807. = Leilus Swainson, 1833; Zool. Illustr. (2): pls. 125–126, 129–130. Unnecessary replacement name for Urania Fabricius, 1807. = Dasycephalus Walker, 1854; List Spec. Lepid. Insects Colln Br. Mus. 1: 4. Unavailable name. = Uranidia Westwood, 1879; Trans. Zool. Soc. Lond. 10: 516, 520–52. Unnecessary replace- ment name for Urania Fabricius, 1807. = Lelius; Swainson, 1836, Nat. Hist. Classif. Birds 1: 233. Misspelling. 1. Urania boisduvalii Guérin-Meneville,1829 Urania boisduvalii Guérin, 1829, Icon. Ins., p. 490. Type locality: Cuba. Type lost. = Urania fernandinae MacLeay, 1837; Trans. Zool. Soc. Lond. 1: 180. Type locality: Cuba. 2. Urania fulgens Walker, 1854 Urania fulgens Walker, 1854, List Lep. Het. Br. Mus. 1, p. 5. Type locality: “Sta. Fe de Bogota”, “Mexico”. Syntypes in BMNH. Walker (1854) lists two specimens “from Mr. Argent’s collection”; Mexican specimen missing (A. Zilli, pers. comm.). = Cydimon cacica Guenée, 1857, Spec. génér. Lep. 9(9): 8. Type locality: “Mexique: Environs d’Acapulco”. 2a. Urania fulgens poeyi (Herrich-Schäffer, 1866) revisedstatus

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 9 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

Cydimon poeyi Herrich-Schäffer, 1866, Corr. Blatt. Ver. Regensb. 20, p. 135. Type locality: Cuba. Type not located; possibly in Gundlach collection, Havana (currently inaccessible). 3. Urania leilus (Linnaeus, 1758) Papilio leilus Linnaeus, 1758, Syst. Nat. 10: 462. Type locality: America. Type in Uppsala University Zoological Museum, Sweden. = Papilio leilaria Hübner, [1807], Verz. Bek. Schmett.: 289. Unjustified emendation of Papi- lio leilus Linnaeus, 1758. = “leilus” surinamensis Swainson, 1833, Zool. Illustr. (2)3: pl. 125. Type locality: [Surinam]. = Cydimon amphiclus Guenée, 1857, Spec. génér. Lep. 9(9): 7. Type locality: Trinidad. = Urania curvata Pfeiffer, 1917, Ent. Z. 31: 70. Type locality: “Puerto Patinm [sic] in Bolivien”. = Urania leilus ab. elegans Niepelt, 1930, Ent. Z. 44: 18. Type locality: . = Urania leilus var. intermedia Pfeiffer, 1917 (in litt.). 3a. Urania leilus brasiliensis (Swainson, 1833) revised status Leilus brasiliensis Swainson, 1833, Zool. Ill. (2)3, pl. 126. Type locality: “Brazil”. Type in BMNH. 4. Urania sloanus (Cramer, 1776) Papilio sloanus Cramer, 1776, Pap. Exot. p.134. Type locality: “Jamaika”. Type in BMNH. = Leilus occidentalis Swainson, 1833, Zool. Ill. 2(3), pl. 129. Type locality: Jamaica. = Urania sloanaria Hübner, 1816, Verz. Bek. Schmett.: 289. Type locality: Jamaica. Unjusti- fied emendation of Papilio sloanus Cramer, 1779. = Urania sloaneus Herbst (misspelling).

Supporting Information S1 Table. Material examined, localities and GenBank accessions. (XLS) S2 Table. List of non-tailed primers used in Sanger sequencing to amplify six short, over- lapping fragments of the COI-5P barcode region. (XLSX) S3 Table. Detail on the technologies used—Sanger or NGS—to successfullyamplify each fragments (F1-F6) of COI-5P barcode region. To see additional information on primer com- binations, reactions attempted on each sample, and download the trace files please go to http:// dx.doi.org/10.5883/DS-URANIA. (XLSX) S4 Table. Min, max, and average base coverage for 10 samples amplified using NGS post- QC read filtering. Although the minimum coverage was 0 for all samples (because none of the samples had 100% success), for the final consensus sequence we generally only included regions with a minimum coverage of 10, unless we were absolutely certain that the low-cover- age reads were accurate. (XLSX)

Acknowledgments We are very grateful to Carmen Pozo and Blanca Prado of the Chetumal node of the Mexican Barcode of Life (Mexbol) network, Yves Basset (Panama), Erik van Nieukerken (The Nether- lands), Rodolphe Rougerie and Bernard Dardenne (France), Eric van Schayck, Ronald Bechlin, Axel Hausmann and Egbert Friedrich (Germany) and Marko Mutanen (Finland) for permission

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 10 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

to include their sequences in this study, as well as to James Hayden (McGuire Center for Lepi- doptera, Gainesville, Florida, USA) for a loan of Urania specimens. Rayner Nuñez (Cuba), Alberto Zilli (UK), Joël Minet (France) and Wolfram Mey (Germany) helped in locating some of the type material of Urania. We also thank David Lees, Pasi Sihvonen and Dorothee Huchon for their reviews.

Author Contributions Conceptualization: VN BCS. Data curation: VN SP. Formal analysis: VN SP BCS. Funding acquisition: PDNH. Investigation: VN BCS. Methodology:VN BCS SP. Project administration: VN. Resources: VN PDNH. Software: VN SP. Supervision: VN PDNH. Validation: VN PDNH. Visualization: VN. Writing – original draft: VN. Writing – review & editing: VN BCS SP PDNH.

References 1. Lewis CB (1944) The Jamaica Emperor. Natural History Society of Jamaica 2: 133. 2. Perkins L (1944) Urania sloanus. Natural History Society of Jamaica 2: 144. 3. Gosse PH (1881) Urania sloanus at home: II. The larva and pupa. Entomologist 14: 241±245. 4. Hemsley WB (1885) Report of the scientific results of the voyage of HMS Challenger during the years 1873±76. Botany, 1 (3). Reports on the botany of Juan Fernandez, the south-eastern Moluccas, and the Admiralty Islands. Her Majesty's Stationery Office, London. 333 pp. 5. Adams CD (1972) Flowering plants of Jamaica. University of the West Indies, Mona, Jamaica. 848 pp. 6. Cramer P (1775±1780) De uitlandsche Kapellen voorkomende in de drie Waereld-Deelen Asia, Africa en AmericaÐPapillons exotiques des trois parties du monde l'Asie, l'Afrique et l'Amerique. Amstel- dam, Chez S. J. Baalde, 1775±1780. 7. Sloane H (1725) A voyage to the islands Madera, , Nieves, S. Christophers and Jamaica [. . .], Vol. II. London: British Museum. 8. Gosse PH (1851) A naturalist's sojourn in Jamaica. London: Longman, Brown, Green and Longmans. 572 pp. 9. Gosse PH (1880) Urania sloanus at home. Entomologist 13: 133±135. 10. Townsend CHT (1893) Swarmings of Urania sloanus. Journal of the Institute of Jamaica 1: 379. 11. Lees DC, Smith NG (1991) Food plant associations of the Uraniinae (Uraniidae) and their Systematic, evolutionary, and ecological significance. Journal of the Lepidopterists' Society 45: 296±347. 12. Vinciguerra R (2009) Osservazioni su Urania sloanus (Cramer, 1779) (Lepidoptera: Uraniidae). SHI- LAP Revista lepidopterologica 37: 307±312.

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 11 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

13. Kendall RO (1978) Periodic occurrence of Urania fulgens (Uraniidae) in the United States. Journal of the Lepidopterists' Society 32: 307±309. 14. Kirby WF (1897) A Handbook to the Order Lepidoptera. Vol. III: Butterflies. London: Edward Lloyd, 308 pp. 15. Leussler RA (1918) Interesting Occurrences at Beeville, Texas. Entomological News 29: 149. 16. Lees DC (1992) Foreleg stridulation in male Urania moths (Lepidoptera: Uraniidae). Zoological Journal of the Linnean Society 106: 163±170. 17. Domgaøa P, Larysz A, Dobosz R, GonzaÂlez JM (2015) Urania sloanus (Cramer, 1779), an extinct spe- cies in the collection of the Upper Silesian Museum (Muzeum GoÂrnośląskie), Bytom, Poland (Lepidop- tera: Uraniidae). SHILAP Revista de Lepidopterologica 43: 455±460. 18. Gaede M (1930) Urania. In: Seitz A, The Macrolepidoptera of the World. 6: 820±831. 19. Smith NG (1992) Reproductive behavior and ecology of Urania (Lepidoptera: Uraniidae) moths and of their larval food plants, Omphalea spp (Euphorbiaceae). Pp. 576±593 In: Quintero DA, Aiello A (eds.) Insects of Panama and Mesoamerica Selected studies. Oxford University Press, Oxford. 20. Wahlberg N, Wheat CW, Peña C (2013) Timing and patterns in the taxonomic diversification of Lepi- doptera (Butterflies and Moths). PLOS one 8: e80875. doi: 10.1371/journal.pone.0080875 PMID: 24282557 21. Mutanen M, Wahlberg N, Kaila L (2010) Comprehensive gene and taxon coverage elucidates radiation patterns in moths and butterflies. Proceedings of the Royal Society of London B Biological Sciences 277: 2839±2848. 22. HeikkilaÈ M, Mutanen M, Wahlberg N, Sihvonen P, Kaila L (2015) Elusive ditrysian phylogeny: an account of combining systematized morphology with molecular data (Lepidoptera). BMC Evolutionary Biology 15: 260, doi: 10.1186/s12862-015-0520-0 PMID: 26589618 23. Lees DC (1992) Foreleg stridulation in male Urania moths (Lepidoptera: Uraniidae). Zoological Journal of the Linnean Society 106: 163±170. 24. Barro A, Vater M, PeÂrez M, Coro F (2009) Surface structure of sound emission organs in Urania moths. In Gorb SN, ed. Functional Surfaces in Biology, Little Structures with Big Effects, Volume. 10. Springer Science Business Media BV doi: 10.1007/978-1-4020-6697-9_11 25. PeÂrez AÂ lvarez M, Barro AC (2014) Funcionalidad del oÂrgano timpaÂnico de Urania boisduvalii (Lepidop- tera: : Uraniidae) y su posible importancia conductual. Revista Cubana de Ciencias Bio- loÂgicas 3: 81±94. 26. Kite GC, Scofield AM, Lees DC, Hughes M, Smith NG (1997) Alkaloidal glycosidase inhibitors and digestive glycosidase inhibition in specialist and generalist herbivores of Omphalea diandra. Journal of Chemical Ecology 23: 121±135. 27. Srygley RB, Dudley R (2008) Optimal strategies for insects migrating in the flight boundary layer: Mechanisms and consequences. Integrative and Comparative Biology 48: 119±133. doi: 10.1093/icb/ icn011 PMID: 21669778 28. Holloway JD (1998) The Moths of Borneo: families Castniidae, Callidulidae, Drepanidae and Uranii- dae. Malayan Nature Journal 52: 1±155. 29. Lafontaine JD (2004) The moths of North America. Noctuoidea: Noctuidae (part). Noctuinae (partÐ Agrotini). Fasc. 27.1. In: Dominick RB, Ferguson DC, Franclemont JG, Hodges RW, Munroe EG (Eds) The moths of North America. Wedge Entomological Research Foundation, Washington DC. 385 pp. 30. Prosser SWJ, DeWaard JR, Miller SE, Hebert PDN (2016) DNA barcodes from century-old type speci- mens using next-generation sequencing. Molecular Ecology Resources 16: 487±497. doi: 10.1111/ 1755-0998.12474 PMID: 26426290 31. Lees DC, Rougerie R, Zeller-Lukashort C, Kristensen NP (2010) DNA mini-barcodes in taxonomic assignment: a morphologically unique new homoneurous moth clade from the Indian Himalayas described in Micropterix (Lepidoptera, Micropterigidae). Zoologica Scripta 39: 642±661. 32. Li H, Durbin R (2009) Fast and accurate short read alignment with Burrows-Wheeler transform. Bioin- formatics 25: 1754±1760. doi: 10.1093/bioinformatics/btp324 PMID: 19451168 33. Tamura K, Peterson D, Peterson N, Stechrer G, Nei M (2011) MEGA5: Molecular evolutionary genet- ics analysis using maximum likelihood, evolutionary distance and maximum parsimony methods. Molecular Biology and Evolution 28: 2731±2739. doi: 10.1093/molbev/msr121 PMID: 21546353 34. Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52: 696±704. PMID: 14530136

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 12 / 13 Phylogeny of Urania (Lepidoptera: Uraniidae)

35. Ronquist F, Teslenko M, van der Mark P, Ayres D, Darling A, H'ohna S, et al. (2012) MrBayes 3.2: Effi- cient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biol- ogy 61: 539±542. doi: 10.1093/sysbio/sys029 PMID: 22357727 36. Templeton AR, Routman E, Phillips CA (1995) Separating population structure from population history: a cladistic analysis of the geographical distribution of mitochondrial DNA haplotypes in the tiger sala- mander, Ambystoma tigrinum. Genetics 140: 767±782. PMID: 7498753 37. Ronquist F (1996) DIVA (dispersal±vicariance analysis) ver. 1.2. Computer program and manual avail- able by anonymous FTP from Uppsala University. (ftp.uu.se or ftp.systbot.uu.se). Accessed March 2016. 38. McLoughlin S (2001) The breakup history of Gondwana and its impact on pre-Cenozoic floristic provin- cialism. Australian Journal of Botany 49: 271±300. 39. Smith NG (1983) Host plant toxicity and migration in the dayflying moth Urania. Florida Entomologist 66: 76±85. 40. Dudley R, Srygley RB, Oliveira EG, DeVries PJ (2002) Flight speeds, lipid reserves, and predation of the migratory Neotropical moth Urania fulgens (Uraniidae). Biotropica 34: 452±458. 41. van den Heyden T (2006) Eine Massenwanderung von Urania fulgens Walker, 1854 in Costa Rica, Zentralamerika (Lepidoptera: Uraniidae). Nachrichten des Entomologischen Vereins Apollo (NF) 27: 55±57. 42. Murillo-Hiller LR (2008) Notas sobre el comportamiento y la migracioÂn de Urania fulgens (Lepidoptera: Uraniidae) en Costa Rica. Acta ZooloÂgica Mexicana 24: 239±241. 43. Sane SP, Srygley RB, Dudley R (2010) Antennal regulation of migratory flight in the neotropical moth Urania fulgens. Biology Letters 6: 406±409. doi: 10.1098/rsbl.2009.1073 PMID: 20181558 44. Catala R (1940) Variations expeÂrimentales de Chrysiridia madagascariensis Less. (Lep. Uraniidae). PhD Thesis. Archives du MuseÂum National D'Histoire Naturelle 17. 45. Griveaud P (1959) Sur les deplacements de Chrysiridia madagascariensis (Lepidoptera Uranidae). Naturaliste Malgache Tananarive 11: 107±109. 46. Montes C, Cardona A, Jaramillo C, Pardo A, Silva JC, Valencia V, et al. (2015) Middle Miocene closure of the Central American Seaway. Science 348: 226±229, doi: 10.1126/science.aaa2815 PMID: 25859042 47. Gregory-Wodzincki KM (2000) Uplift history of the Central and Northern Andes: A review. Geological Society of America Bulletin 112: 1091±1105. 48. Drake VA, Farrow RA (1988) The influence of atmospheric structure and motions on migration. Annual Review of Entomology 33: 183±210. 49. Viette P (1972) Contribution à l'etude des lepidopteres Uraniidae Malgaches. Bulletin mensuel de la SocieÂte linneÂenne de Lyon 41: 189±193. 50. Emmel TC, Farkas VJ (1974) Urania fulgens (Uranidae) captured in Florida. Journal of the Lepidopter- ists' Society 28: 292. 51. Buskirk RE (1985) Zoogeographic patterns and tectonic history of Jamaica and the northern Carib- bean. Journal of Biogeography 12: 445±461.

PLOS ONE | DOI:10.1371/journal.pone.0164405 October 20, 2016 13 / 13